Spectral-DP: Differentially Private Deep Learning through Spectral Perturbation and Filtering
Engineers explore architecture and security mechanisms for Spectral-DP: Differentially Private Deep Learning through Spectral Perturbation and Filtering.
Summary
No summary yet — watch the video above, or open it on YouTube.
Related talks
IEEE Symposium on Security and Privacy·23 min
Presenters detail vulnerability discovery and exploitation in Differentially Private Model Publishing For Deep Learning.
deep-learningdifferentiallyprivate+3
IEEE Symposium on Security and Privacy·19 min
Analysts investigate attack surfaces and defenses concerning Hark: A Deep Learning System for Navigating Privacy Feedback at Scale.
privacydeep-learninghark+3
IEEE Symposium on Security and Privacy·3 hr 29 min
Analysts investigate attack surfaces and defenses concerning [DLS2021] 4th Deep Learning And Security Workshop -- Part 2.
deep-learningworkshopdls2021+3